Classification/clustering issue (supervised and unsupervised)

[jelrod14]

Hi,
Can't seem to get either clustering method to work, and both result in this error after selecting the detection file:

Scalar structure required for this assignment.

Error in CreateClusteringData (line 60)
        Calls_tmp.audiodata_index = repmat(j, height(Calls_tmp), 1);

Error in UnsupervisedClustering_Callback (line 25)
                            [ClusteringData, ~, ~, ~, spectrogramOptions] = CreateClusteringData(handles, 'forClustering', true, 'save_data', true);

Error in gui_mainfcn (line 95)
        feval(varargin{:});

Error in DeepSqueak (line 29)
    gui_mainfcn(gui_State, varargin{:});

Error in matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)DeepSqueak('UnsupervisedClustering_Callback',hObject,eventdata,guidata(hObject))
 
Error while evaluating Menu Callback.

I've tried restarting after selecting correct audio and detection folders, as well as loading the calls successfully.
Thanks!

[DrCoffey]

Hey @jelrod14, I just pushed a major update with a lot of bug fixes. Try downloading the new version, and make sure you are using a recent release of MATLAB (2021a or later). Let me know if that fixes your problem, clustering is working well on my end.

[jelrod14]

Gotcha, I'll test it out! Since originally posting, I've figured out that the issue may be that deepsqueak 2.6 stored detections in a data structure file, while 3.0 now utilizes a table format for detection files. I am not sure if there is a way to convert these files or change the structure to make it compatible? my entire dataset (500ish detection files) was generated by a lab member before me in the 2.6 version. Obviously I could redetect in the newer version but would like to avoid this if possible 😅